- Horizon AI
- Posts
- Google DeepMind's New AI Models for Controlling Robots 🤖
Google DeepMind's New AI Models for Controlling Robots 🤖
How to Screen Share with ChatGPT
Gina Acosta
March 13, 2025
In partnership with
Welcome to another edition of Horizon AI,
Google DeepMind is introducing two new AI models, which they hope will lay the foundation for a new generation of helpful AI-powered robots.
Let’s get into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Google Deepmind Introduces New AI Models for Robotic Control
Chinese AI Team Wins Global Award for Using Industrial Chip Instead of Nvidia GPU
AI Tutorial: How to Screen Share with ChatGPT
AI Tools to check out
The Latest in AI and Tech 💡
AI Findings/Resources
AI News
GOOGLE DEEPMIND
Google Deepmind Introduces New AI Models for Robotic Control
Google DeepMind is introducing two new AI models designed to help robots “perform a wider range of real-world tasks than ever before.”
Details:
The first model, called Gemini Robotics, is built on Gemini 2.0 and is a vision-language-action model capable of understanding new situations, even if it hasn’t been trained on them.
The new model makes advancements in three key areas that Google DeepMind says are essential to building helpful robots: generality, interactivity, and dexterity.
The other model, Gemini Robotics-ER (short for “embodied reasoning”), is an advanced visual-language model that the company says can “understand our complex and dynamic world.”
It combines spatial reasoning with Gemini’s coding abilities to create new capabilities in real time. For example, when interacting with a coffee mug, the model can determine an appropriate two-finger grasp for picking it up by the handle and calculate a safe trajectory for approaching it.
Google continues to push the boundaries of AI and robotics to “build the next generation of humanoid robots” that are adaptable and capable of assisting in real-world scenarios.
TOGETHER WITH BELAY
Accomplish More. Juggle Less.
When you love what you do, it can be easy to take on more — more tasks, more deadlines, more hours – but before you know it, you don’t have time to do what you loved in the beginning. Don’t just do more – do more of what you do best.
BELAY’s flexible staffing solutions leverage industry experience with AI systems to increase productivity without sacrificing quality. You can accomplish more and juggle less with our exceptional U.S.-based Virtual Assistants, Accounting Professionals, and Marketing Assistants. Learn how with our free ebook, Delegate to Elevate, and leave the more to BELAY.
AI RESEARCH
Chinese AI Team Wins Global Award for Using Industrial Chip Instead of Nvidia GPU
Researchers successfully developed the first FPGA-trained video-generation AI model on an off-the-shelf industrial chip, surpassing high-end GPUs in both speed and efficiency.
Details:
A Chinese team from Shanghai Jiao Tong University, Tsinghua University, and Infinigence-AI trained their FlightVGM video-generation model using AMD's widely available V80 FPGA chip.
FlightVGM achieved a 30% performance boost and was 4.5 times more energy efficient than Nvidia’s RTX 3090 GPU.
The innovation earned the Best Paper Award at FPGA 2025, marking the first win for a mainland Chinese team.
This approach offers a new alternative for industries to deploy cost-effective, energy-efficient AI systems, from robotic controls to autonomous vehicles.
AI Tutorial
How to Screen Share with ChatGPT
If you've ever wanted ChatGPT to guide you through a task without constant back-and-forth text, its screen-sharing feature lets you do just that, providing real-time assistance on any project.
Note: To use ChatGPT's screen-sharing feature, you'll need a ChatGPT Plus subscription ($20/month) and the ChatGPT mobile app.
Download the ChatGPT app for Android or iOS and open it to start a new conversation.
Tap the voice mode icon in the bottom right corner to access advanced voice mode.
Tap the three dots and select Share Screen.
When prompted, grant ChatGPT permission to access your screen.
Choose to share your entire screen or a specific app, then tap Start.
AI Tools to check out
💼 Harvey: Domain-specific AI for law firms, professional service providers, and the Fortune 500.
🗣 Wispr Flow for Windows: Use your voice to write 3x faster in every application.
✅ Greta: Ship any full-stack applications within seconds without writing a single line of code.
🧊 Quadratic: Chat with your data and get insights in seconds.
✨ Whisper Large V3: Transcribe long-form microphone or audio inputs with the click of a button
AI Findings/Resources
📚 All the FREE IBM Data Science courses + certificates
🤔 5 reasons why AI search tools might never fully replace classic Google searches
👀 Is AI a bubble? It's complicated
🧠 How to learn anything 10x faster than anyone with AI
The latest in AI and Tech
Anthropic's CEO warns that spies are targeting $100M AI secrets hidden in just a ‘few lines of code’
Dario Amodei is asking the U.S. government to step in to defend the country’s top AI companies against foreign spies, especially from China, who are trying to obtain costly “algorithmic secrets.”
Snapchat Platinum subscribers will now be able to use video lenses featuring AI-generated animals or flowers. Initially, there will be three of these “AI Video Lenses”: the fox lens, the raccoon lens, and the flower lens.
In a proposal for the U.S. government’s “AI Action Plan,” OpenAI called for a U.S. copyright strategy that “[preserves] American AI models’ ability to learn from copyrighted material.
Operator's research preview is now available to those 18 and older in Europe, Norway, Iceland, Liechtenstein, and Switzerland at operator.chatgpt.com. The feature is only available to Pro users for now.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻